LSTM-Based Predictions for Proactive Information Retrieval

Petri Luukkonen, Markus Koskela, Patrik Floreen

Research output: Chapter in Book/Report/Conference proceedingConference contributionProfessional

Abstract

We describe a method for proactive information retrieval targeted at retrieving relevant information during a writing task. In our method, the current task and the needs of the user are estimated, and the potential next steps are unobtrusively predicted based on the user's past actions. We focus on the task of writing, in which the user is coalescing previously collected information into a text. Our proactive system automatically recommends the user relevant background information. The proposed system incorporates text input prediction using a long short-term memory (LSTM) network. We present simulations, which show that the system is able to reach higher precision values in an exploratory search setting compared to both a baseline and a comparison system.
Original languageEnglish
Title of host publicationUnknown host publication
Number of pages6
Publication date2016
Publication statusPublished - 2016
MoE publication typeD3 Professional conference proceedings
EventAdvances in neural information processing systems - Pisa, Italy
Duration: 21 Jul 201621 Jul 2016
Conference number: 1 (Neu-IR)

Fields of Science

  • 113 Computer and information sciences

Cite this

Luukkonen, P., Koskela, M., & Floreen, P. (2016). LSTM-Based Predictions for Proactive Information Retrieval. In Unknown host publication
Luukkonen, Petri ; Koskela, Markus ; Floreen, Patrik. / LSTM-Based Predictions for Proactive Information Retrieval. Unknown host publication. 2016.
@inproceedings{20a3952d123d4531a7c1299d4e532e09,
title = "LSTM-Based Predictions for Proactive Information Retrieval",
abstract = "We describe a method for proactive information retrieval targeted at retrieving relevant information during a writing task. In our method, the current task and the needs of the user are estimated, and the potential next steps are unobtrusively predicted based on the user's past actions. We focus on the task of writing, in which the user is coalescing previously collected information into a text. Our proactive system automatically recommends the user relevant background information. The proposed system incorporates text input prediction using a long short-term memory (LSTM) network. We present simulations, which show that the system is able to reach higher precision values in an exploratory search setting compared to both a baseline and a comparison system.",
keywords = "113 Computer and information sciences",
author = "Petri Luukkonen and Markus Koskela and Patrik Floreen",
note = "Volume: Proceeding volume:",
year = "2016",
language = "English",
booktitle = "Unknown host publication",

}

Luukkonen, P, Koskela, M & Floreen, P 2016, LSTM-Based Predictions for Proactive Information Retrieval. in Unknown host publication. Advances in neural information processing systems, Pisa, Italy, 21/07/2016.

LSTM-Based Predictions for Proactive Information Retrieval. / Luukkonen, Petri; Koskela, Markus; Floreen, Patrik.

Unknown host publication. 2016.

Research output: Chapter in Book/Report/Conference proceedingConference contributionProfessional

TY - GEN

T1 - LSTM-Based Predictions for Proactive Information Retrieval

AU - Luukkonen, Petri

AU - Koskela, Markus

AU - Floreen, Patrik

N1 - Volume: Proceeding volume:

PY - 2016

Y1 - 2016

N2 - We describe a method for proactive information retrieval targeted at retrieving relevant information during a writing task. In our method, the current task and the needs of the user are estimated, and the potential next steps are unobtrusively predicted based on the user's past actions. We focus on the task of writing, in which the user is coalescing previously collected information into a text. Our proactive system automatically recommends the user relevant background information. The proposed system incorporates text input prediction using a long short-term memory (LSTM) network. We present simulations, which show that the system is able to reach higher precision values in an exploratory search setting compared to both a baseline and a comparison system.

AB - We describe a method for proactive information retrieval targeted at retrieving relevant information during a writing task. In our method, the current task and the needs of the user are estimated, and the potential next steps are unobtrusively predicted based on the user's past actions. We focus on the task of writing, in which the user is coalescing previously collected information into a text. Our proactive system automatically recommends the user relevant background information. The proposed system incorporates text input prediction using a long short-term memory (LSTM) network. We present simulations, which show that the system is able to reach higher precision values in an exploratory search setting compared to both a baseline and a comparison system.

KW - 113 Computer and information sciences

UR - https://www.microsoft.com/en-us/research/event/neuir2016/

M3 - Conference contribution

BT - Unknown host publication

ER -

Luukkonen P, Koskela M, Floreen P. LSTM-Based Predictions for Proactive Information Retrieval. In Unknown host publication. 2016